Skip to content

// installation

basebox Server

Applies to

Product: Server · Audience: Platform Operator

basebox Server is a dedicated physical server for one customer. Four independent decisions describe every server deployment:

Dimension Options Page
Hardware source FAST LTA / basebox-supplied · procured by the customer Hardware options
Location Customer data center · basebox data center Hosting options
Operation by the customer · by basebox (if commissioned) Operating models
Installation path Bare metal · existing Kubernetes Installation paths

None of them changes the product: a dedicated customer server hosted at basebox is still Server, not Cloud. How the server is built technically: Server architecture.

In this section

  1. Decide: the four dimensions above – usually settled within an hour.
  2. Understand: Understand the architecture and Server architecture – why GPUs serve the service models and inference, not basebox.
  3. Check hardware: against the Reference configurations.
  4. Install: Bare-metal installation or directly Using Helm charts.
  5. Choose and connect a model: Models & inference.
  6. Accept: Validate installation.
  7. Operate: Operations.

What you will not find here

Application administration (users, apps, enabling connectors) is under Administration; API usage under Developer → API; security properties per hosting and operating model under Security & Compliance → Server. A non-technical entry point is Start with basebox Server.